MZ-TH/10-47
1012.4600 [hep-ph]
December 2010
A survey of top quark polarization
at a polarized linear e + e~ collider
S. Groote
Loodus- ja Tehnoloogiateaduskond, Fiiiisika Instituut,
Tartu Ulikool, Riia 142, EE-51014 Tartu, Estonia
J.G. Korner
Institut fur Physik der Johannes- Gutenberg- Universitat,
Staudinger Weg 7, D-55099 Mainz, Germany
B. Melic
Rudjer Boskovic Institute, Theoretical Physics Division,
Bijenicka c. 54, HR-10000 Zagreb, Croatia
S. Prelovsek
Physics Department at University of Ljubljana
and Jozef Stefan Institute, SI-1000 Ljubljana, Slovenia
Abstract
We discuss in detail top quark polarization in above-threshold (tt) production
at a polarized linear e + e~ collider. We pay particular attention to the minimization
and maximization of the polarization of the top quark by tuning the longitudinal
polarization of the e + and e~ beams. The polarization of the top quark is cal-
culated in full next-to-leading order QCD. We also discuss the beam polarization
dependence of the longitudinal spin-spin correlations of the top and antitop quark
spins.
1
1 Introduction
A future linear e + e~ collider offers the cleanest conditions for studying top quark prop-
erties, such as the top quark mass, its vector and axial couplings, and possible magnetic
and electric dipole moments. Apart from these static properties, also the polarization
of the top quark can be studied with great precision. The top decays sufficiently fast so
that hadronization effects do not spoil the polarization which it has at its birth. The
large number of top quark pairs expected to be produced at the ILC, e.g., 50 (ti)/hour
at 500 GeV (based on a luminosity of L = 2 x fO^cm" 2 ^" 1 [TJ [2]), will enable one to
precisely determine the top quark polarization from an angular analysis of its decay
products in the dominant decay t — > Xj, + W + . The expected statistical errors in the
angular analysis are below the 1% level. Therefore very precise measurements of the an-
gular distributions and correlations of the decay products of t and t will shed light on the
polarization of the top quarks and on the spin-spin correlations of the top and antitop
quark pairs which are imprinted on the top and antitop quarks by the (tf)-production
mechanism. In addition, the measurement of the top polarization will make it possible
to precisely determine the electroweak Standard Model (SM) parameters or to study a
variety of new phenomena beyond the SM.
It is well known that the top quarks from e + e~ annihilations are polarized even
for unpolarized e + e~ beams due to the presence of parity-violating interactions in the
Standard Model (SM). One also knows from the work of Ref. pi H] that the polarization
of the top quark in polarized e + e~ annihilations can become quite large when the beam
polarization is adequately tuned. This is illustrated in Fig.[T]where we display the energy
dependence of the mean longitudinal and transverse polarization of the top quark in the
helicity system for different values of the effective polarization P e ff defined by
P h -~ h +
-r eff
I - h-hi
In Eq. (JTJ), h- and h + are the longitudinal polarization of the electron and positron beams,
respectively. Note that, for unpolarized positron beams h + = 0, one has P e & = h_. For a
given value of h-, even small values of positron polarization of opposite sign will enhance
the effective beam polarization. We shall return to this point in Sec. 2. As compared to
the work of Ref. [3], Fig. CD now includes the 0(a s ) radiative corrections. Large single-
spin polarization effects due to beam polarization effects are also implicit in the work of
Parke and Shadmi [5] . Although Ref. [5] is designed for the analysis of top-antitop quark
spin correlations, it is easily adapted to single-spin polarization effects as also discussed
in Ref. ©.
We shall see that the polarization of the top quark is governed by three parameters:
the velocity v = y 1 — 4m 2 /s, the effective polarization P e g, and the cosine of the scat-
tering angle cos 6. At the respective boundaries of the three parameters the description
2
1 Q Q U I I I I I I I i i i i I I I I I I I i i i I I I I I I I I L
400 500 600 700 800 900 1000
Vs [GeV]
(a)
CL
00
50
-50
n i i i i r
n i i i i r
Peff
P.ff = +•
n i i i i i i r
1 Q Q U i I i i i i I i i i i I i i i i I i i i i I i i i i I i i l
400 500 600 700 800 900 1000
Vs [GeV]
(b)
Figure 1: Average (a) longitudinal polarization (P^) and (b) transverse polarization
(p(* r )} as a function of the cm. energy y/s, for the values P cS = —1 (dashed), P eff = —0.5
(dash-dotted), P eS = (solid), P cS = +0.5 (dash-dotted), and P cff = +1 (dashed).
Averaging is over cos 9.
of the polarization phenomena becomes reasonably simple, in particular, at the Born
term level. The limits v = (threshold) and v = 1 (high-energy limit) are discussed in
Sec. 4. In Sec. 5 we discuss the limiting cases P c s = ±1.
The two respective limiting cases v — 0, 1 and P ff = ±1 contain in a nutshell
much of the information that we want to discuss in the remaining part of the paper
for intermediate values of these parameters. Many of the qualitative features of our
results can be understood from extrapolations away from the two respective limits.
We shall also address the question of how to maximize and minimize the polarization
of the top quark by tuning the beam polarization. Whereas a maximum polarization is
optimal for the experimental determination of polarization effects, it is often desirable
to gauge the quality of a polarization measurement against the corresponding unpolar-
ized decay analysis. For some measurements it may even be advantageous to eliminate
polarization effects altogether.
Of course, in the tuning process one has to bear in mind to keep the production
rate at an acceptable level. This problem is not unrelated to the one of the original
motivations of including beam polarization in linear colliders, namely, the gain in rate
through beam polarization effects. We shall also address this question.
Our paper is structured as follows. In Sec. 2 we present the spin formalism of polar-
ized beam production of top-antitop quark pairs including the polar angle dependence of
the various spin components and longitudinal beam polarization effects. We present Born
term and loop formulas for the relevant structure functions and collect general expres-
sions necessary for the numerators and the denominator of the polarization observables.
Section 3 contains numerical next-to-leading (NLO) results on the angle-integrated rate
and on polar angle distributions of the rate including beam polarization effects. We also
provide numerical results on the left-right polarization asymmetry Alr. In Sec. 4 we
discuss the limiting cases v — and v = 1 at the Born term level. In Sec. 5 we describe
the simplifications that occur for maximal effective beam polarizations P e g = =Fl which
correspond to the (e7, K , e^ L ) beam configurations. In Sec. 6 we discuss beam polar-
izations effects on the three components of the top quark polarization vector. Section 7
contains a discussion of the magnitude and the orientation of the polarization vector of
the top quark. In Sec. 8 we present numerical NLO results on beam polarization ef-
fects on longitudinal spin-spin correlations of the top and antitop quark. Finally, Sec. 9
contains a summary of our results and our conclusions. In an Appendix we list the elec-
troweak coupling coefficients used in this paper and relate them to the chiral electroweak
coupling coefficients used e.g. in Ref. [5].
Many of the quantitative arguments presented in this paper are based on Born term
level results for which we give explicit alpha-numerical expressions for y/s = 500 GeV.
We emphasize, though, that all numerical results presented in the plots include the full
0(a s ) radiative corrections where we have integrated over the full gluon phase space.
4
By comparing the graphical NLO results with the numerical LO results, one can assess
the size of the 0(a s ) radiative corrections, at least for the representative energy of
y/s = 500 GeV. In general, the 0(a s ) corrections to polarization observables are small
(up to several percent) but can become much larger in some areas of phase space. A
case in point is the longitudinal polarization of the bottom quark produced on the Z°
at the backward point which obtains a 25% 0(a s ) correction when P e g = +1 jl]. As we
shall see later on, the 0(a s ) corrections to (tfj-production can amount up to 12% (see
Sec. 7). In addition, there are polarization observables that are zero at the Born term
level and become populated only at 0(a s ). Among these are the normal component of the
polarization (see Sec. 6) and the longitudinal polarization produced from a longitudinal
intermediate vector boson (see Sec. 2).
2 Spin formalism of polarized beam production
The production of top quark pairs at a linear e + e~-collider proceeds via 7- and Z-
exchange,
e~e +1 4ti. (2)
At the center of mass energies which are being envisaged at the ILC (s = {p e - +p e +) 2 ),
a/s ~ 2m t -T- 1000 GeV, top quark pairs will be produced with nonrelativistic velocities
in the threshold region (v — > 0) up to relativistic velocities of v = 0.937 at the highest
energy y/s = 1000 GeV (v = \Jl — 4m 2 /s)[j] This enables the study of the complete
production phenomena with different polarization and correlation effects that reach from
the nonrelativistic to the relativistic domain. For unpolarized beams the total rate is
dominated by the diagonal (7 - 7) and the [Z - Z) rates which contribute at the same
order of magnitude. The (7 - Z) interference contribution to the total rate is suppressed
due to the smallness of the vector [Z e + e~) coupling vi (vi = — 1 + 4 sin 2 0w)- The (7 -
Z) interference contribution can, however, become quite sizable for polarized beams, for
the polar angle dependent rates and for top quark polarization effects.
We mention that, at threshold, there will be the opportunity for very precise mea-
surements of the top quark mass and width, as well as of the strong coupling a s . In this
region, perturbative QCD is no longer applicable. One has to solve the Schrodinger equa-
tion for the relevant Green functions in a nonrelativistic approximation for a Coulombic
potential, i.e. the nonrelativistic QCD (NRQCD) method, described first in Ref. j8] and
later applied to the calculation of various different quantities at threshold (see for exam-
ple the discussion in Ref. [9j [10] and references therein) . In this paper we shall discuss
top-antitop production well above threshold where perturbation theory can be safely ap-
plied. For our purposes we take the perturbative regime to start approximately 10 GeV
1 In the first stage of the ILC, one will reach energies up to 500 GeV with an optional second stage
upgrade to 1000 GeV [HE]- For the multi-TeV collider CLIC one foresees energies up to 3TeV [7].
5
above threshold. Throughout this paper we shall take the top quark mass to have a
nominal value of 175 GeV. Therefore, we shall consider cm. beam energies starting from
y/s = 360 GeV.
We are going to discuss the most general case of the polarization of the top quark
with arbitrary longitudinal polarizations of the e~- and e + -beams. The rate depends
on the set of four parameters G [—1, 1], h + G [—1, 1], v G [0, 1], cos# G [—1, 1]} or,
equivalently, on the set {Kg G [0,2], P c s G [—1, 1], v G [0, 1], cos9 G [—1, 1]}, where we
shall call Kq = 1 — h_h + the gain factor. We have indicated the range of the parameter
values in square brackets. In contrast to the rate, the polarization of the top quark
depends only on the set of the three parameters {P e s, v , cos#}. When discussing our
predictions we shall attempt to explore the whole four- and three-dimensional parameter
space for the rate and polarization, respectively. We mention that the beam polarizations
envisaged at the ILC are h_ = ±90% for electrons and h + = ±80% for positrons [TTj .
We will see that beam polarizations significantly influence the polarization phenom-
ena of a top quark. In addition, adequately tuned beam polarization can enhance the
top-antitop quark signal and suppress other background processes such as IV-pair pro-
duction (see discussion in Ref. [12]).
In what follows, we concentrate on the polarization of the top quark, i.e. we sum
over the polarization of the antitop quark. The polarization of the antitop quark can
be obtained from the corresponding polarization components of the top quark using CP
invariance as will be discussed in the summary section. Even more structure is revealed
when one considers joint top-antitop polarization. In order to reveal this structure, one
must perform a joint analysis of the decay products of the top and antitop quark, (tt)
spin-spin correlations will be briefly discussed in Sec. 8 at the end of the paper.
The
given by
eneral expression of the cross section for (tt) production in e + e collisions is
e 4
da {m) = 2tt- Y, gijL ilxu H^dPS . (3)
S i,3=l
L % is the lepton tensor, Hi is the hadron tensor encoding the hadronic production
dynamics, dPS is the phase space factor, and the are the elements of the electroweak
coupling matrix which are defined in the Appendix. The sum runs over the four inde-
pendent configurations of products of the vector and axial vector currents, i.e. i, j = 1,2
for (VV ± AA)/2, i,j = 3 for i(VA - AV)/2, and i,j = 4 for (VA + AV)/2 for the
product of lepton and quark currents, and m denotes one of the possible polarization
configurations of the top quark: longitudinal [m = £), transverse (m = tr) in the beam
scattering plane and normal (m = n) to the beam scattering plane. Our choice of the
2 The spin kinematics of e + e collisions has been formulated in a number of papers. These include
the unpublished DESY report |13j of which the portions relevant to this paper have been summarized
in Ref. [H]. Other papers on the subject are Refs. [2 H2 [HJ HQ [HI Q3] .
6
Figure 2: A generic configuration for top pair production and top polarization at a
polarized e + e~ collider with a (e£,e^) polarization. The positive z axis points into
the direction of the electron momentum. The angle a is the polar angle of the top
quark polarization relative to the top quark momentum measured anticlockwise from
the direction of the top quark.
three orthonormal spin directions (e^ tr \ e^ n \ e^') are given by
g (tr) = (Pe- X Pt) X Pt £(„) = Pe- X g = fh_ ,^
\(Pe- X Pt) X Pt\ ' \Pe~ X PtY \Pt\ '
In Fig. [2] we have drawn the directions of and for a generic top quark direction;
the vector e^ n ' shows out of the plane. For the unpolarized top quark case the superscript
(m) is dropped in Eq. (J3J). The explicit definitions for all the above quantities together
with explicit analytical expressions for the radiative corrections can be found in Refs. [HI
□H 1201 EH] (see also Ref. [1]).
We proceed with the discussion in the helicity basis, i.e. we take the direction of the
top quark to define the z direction of the hadronic system. For unpolarized beams the
angular decomposition of the differential polarized cross section can be written as
= sin 9 cos 9af r ' n ^ = sin 9a^ r ' n ^ , (5)
where, at NLO of QCD,
^ = ^jl9i 1 {H 3 a {m \Born) + H^\a s )) for a = U,L,I (6)
63 3=1
and
^ = ^ E ^ (^ 4(m) (5orn) + iJ a 4 ( m ) (a.)) for a = F, A . (7)
7
In Eq. ((Sj) we have rewritten the covariant representation ([3]) in terms of helicity structure
functions c^ 71 '. The angle 9 is the polar angle between the momentum of the top quark
and the electron momentum (see Fig. [2]). For example, in the purely electromagnetic
case e + e~ — > 7* — > qq one obtains the LO formula
,UT 27rN c Q 2 f v—(l + cos 2 6 + (1 - v 2 ) sin 2 9
— i/IIV c ^fU— J
cos v As v
using the LO born term expressions listed later in Eq. ffTO]) . The distribution (jSJ) agrees
with Eq. (41.2) in the PDG booklet. We mention that our 0(a s ) corrections agree with
those in Ref. [I] after correcting a sign mistake in the normal polarization (see Erratum
in Ref. [21]).
Above the top quark threshold, one is sufficiently far away from the Z-boson pole
to neglect the imaginary part of the Z boson pole propagator. This can be appreciated
from the Breit-Wigner line shape of the Z propagator, viz.
1 1 / M Z Y Z \ // M|r| \
XZ ~ s-M 2 +iM z T z ~ s- M 2 ^~ % s- MfJA 1+ (s-M 2 yr (9)
The factor M Z T Z / (s — M z ) determines the ratio of the imaginary and real parts of the
Z propagator Imx^/Re^z- It is already quite small at threshold (~ 0.002) and falls off
with s -1 . Dropping the imaginary part contribution of the Z propagator implies that
we neglect contributions proportional to #13 in Eq. ((6]) and #43 in Eq. ([7]). We shall also
neglect the width dependence in the real part of the Z propagator because it is negligibly
small.
The nonvanishing unpolarized Born term contributions H 3 a (Born) read
Hlj(Born) = 2N c s(l + v 2 ), H l L (Born) = iV c s(i - v 2 ) = H 2 L (Born),
H 2 (Born) = 2N c s(l - v 2 ), H^(Born) = 4N c sv. (10)
One has (1 — v 2 ) = 4m 2 /s showing that the longitudinal rate falls off with a s _1
power behaviour relative to the transverse rates Hjjp. The longitudinally polarized
contributions read
H^\Born) = AN c sv, H x /\Born) = 2N c s(l + v 2 ),
H 4 L (e) (Born) = 0, H 2 /\Born) = 2N c s(l - v 2 ). (11)
Note that one has the Born term relations
H^iBorn) = H^(Born) ,
H l p m (Born) = E]j 2 {Born) , (12)
which are due to angular momentum conversation in the back-to-back configuration at
the Born term level. It is quite clear that Eqs. ( f!2l) no longer hold true in general at
8
0(a s ) since quark and antiquark are no longer back-to-back in general due to additional
gluon emission. The relations (fl2~j) will be useful in our subsequent discussion of the
longitudinal polarization at the forward and backward points. Notable also is the relation
Hl {e) (Born) = in Eq. (fU]) which is again related to the LO back-to-back configuration.
The radiative corrections to the corresponding polarization component p£ have been
studied in Ref. [20J and have been found to be small of 0(0.1%) when averaged over
gluon phase space. For small top quark energies Pf 1 can become as large as 3% at
= 500 GeV.
For the transverse polarization components, one has [21]
Hf r \Born) = 2N c sv^L, Hf r \Born) = 2N c s^L = Hf r) (Born) . (13)
V2s V2s
The only nonnegligible contribution to the normal polarization component P^ comes
from the imaginary part of the one-loop contribution (CV = 4/3)
H) {n \loop) = 2N c s^7rv^L = H? n \loo V ) , (14)
4ff v2s
HT\loop) = 2N c s^n(2-v>)^=. (15)
As already mentioned in the Introduction, the transverse and normal polarization compo-
nents can be seen to fall off with a power behaviour of relative to the longitudinal
polarization components.
The a s corrections to the polarized structure functions H^ m ^ = Hl(+s m ) — H 3 a (—s m )
and the unpolarized structure function H 3 a = H a (+s m ) + H a (—s m ) (s m is the polarization
vector of the top quark) are too lengthy to be listed here. They can be found in Refs. [HI
[T9l |20j [21] , or, in a very compact two-page analytical representation, in Sec. 5 of Ref. [22J.
The longitudinal polarization of the electron and positron beams enter the above
formulas as [0
9ij -» {l-h-h+)gij + {h--h+)g4j = (I - h^h + )(g lj + P cS g 4j ) ,
g Aj (1 - h-h+)g4j + (h- - h+)gij = (1 — h-h+){g±j + P^gij) , (16)
where P e ff is defined in Eq. (JTJ). In Eq. ( 1T61) h_ denotes the electron's and h + denotes the
positron's longitudinal polarization which can take values between ±1. An electron with
/i_ = =1=1 will be referred to as the totally polarized left-handed (right-handed) electron
( e L//?)- Similarly, a right-handed positron (e^) has h + = —1 and a left-handed positron
(e£) has h + = +1. From the definition of P e s (see Eq. ([TJ)) it is clear that large values
of P c fi can be reached even for nonmaximal values of h- and h + , as Fig. [3k shows. For
3 Transverse beam polarization effects will not be discussed in this paper because present plans call
for longitudinal beam polarization at the ILC. Transverse beam polarization effects can be included as
described e.g. in Ref. [14] .
9
Figure 3: Contour plots in the h + ) -plane (a) for fixed values of the effective po-
larization P c g = — h + )/(l — h-h + ) and (b) for fixed values of the gain factor
Kg = (1 — h-h+). The two points marked off in the plots correspond to (h^,h + ) =
(-0.8, +0.625) and (+0.8,-0.625), respectively.
10
example, the large value of P e ff = —0.95 can be achieved with h- = —0.8; h + = 0.625,
and correspondingly, P eS = 0.95 can be reached with h_ = 0.8; h + = —0.625. These
two examples have been marked off in Fig. [3l Both sets correspond to a gain factor of
K G = 1.5.
The orient at ion- dependent longitudinal, transverse, and normal polarization compo-
nents which we are interested in are defined by
s\\ da^ m '/d cos 9 „ ,„„ s
P (m) (cos9 = - /, — m = l,tr,n, 17
da I a cost)
where da jd cos 9 is the unpolarized differential cross section. Of course, there is an
additional dependence of the above quantities on the cm. beam energy y/s, and on the
beam polarizations h- and h + to be discussed later on. The unpolarized cross section is
given by the first three terms in Eq. (jSJ) dropping, of course, the label (£).
Dropping the common factor 7ia 2 v/(3s 2 ) in the ratio ( flTl) . we shall represent the
polarization components by the ratios
pM(cosfl) = -A *\ m = £,tr,n. (18)
D{cos9)
In particular, the gain factor Kg has canceled out in the ratio (1T8|) implying that the
polarization only depends on P e fj.
The numerator factors in Eq. ([18]) are given by
+7 cos 9 ( ((741 + 9iiPes)HM + (g 42 + g 12 P cS )H 2 F {e) ) , ( 19 )
N®{cos6) = -(1 + cos 2 9) (g u + g u P cS )Hi w + - sin 2 9 (g li + g i4 P cS )H 4 i ;
3
A
q
N {tr) (cos 0) = r sin 9 cos 9 (g u + g A4 P cS ) Hj {tr)
v2
-y= sin 9 ((^41 + g u P C K)Hf r) + (242 + ffuPrff)^? r) ) , (20)
and by
jV^(cos0) = -A sin # cos # ((2n +24iPc ff )P/ 1(n) (^) + (212 + g 42 P cS ) H 2{n \loop)]
--y= sin (244 + 2i4Pcfr) H 4 A (n) (loop) . (21)
For the denominator, one has
P(cos 9) = -(1 + cos 2 0) ((211 + g4iP cS )Hlj + (212 + 24 2 P e fr)Pf}
+ ^ sin 2 ((211 + g 41 P cS )Hl + (212 + g^P c s)H 2 L
+^cos9(g M + g u P eS )H F . (22)
11
At the forward (FP) and backward (BP) point the transverse and normal polarization
components vanish. Referring to the relations (|T2|) . at Born term level the longitudinal
polarization component P^ takes a very simple form at the forward (FP) and backward
(BP) point for the maximal values of the effective polarization P e s = ±1. One has
FP: P w (cosfl = +l) = ±1,
BP: P w (cosfl = -l) = (23)
in agreement with angular momentum conservation. It is clear that these relations no
longer hold true in general at NLO due to hard gluon emission.
It is useful to define the left-right polarization asymmetry Am through the relation
da(P eS ) - da(-P eS )
da(P cS ) + da(-P eS )
where
A LR
|(1 + cos 2 9) {g 41 H^ + g A2 Hl) + § sin 2 9 (g^Hj + g 42 H 2 L ) + f cos 9g u H 4 F
§(l + cos 2 fl) (g u Hh + g 12 H^j + f sin 2 (s u #i + <? 12 #£) +§cos0(/44#£
(25)
Of interest is the angle a enclosed by the momentum and the polarization of the top
quark projected onto the scattering plane (see Fig. [2])J^| The angle a is determined by
iV (tr) (cos 9)
tan a (cos 0) = ^^ . (26)
Equation (|26|) assumes a simple form at threshold and in the high-energy limit as dis-
cussed in Sec. 4, and for P e ff = ±1 as will be discussed in Sec. 5. The correlations
between a and 9 implied by Eq. (|26|) will be discussed in Sees. 4, 5 and 7.
3 Beam polarization dependence of the rate
We begin our numerical discussion with the rate proportional to the denominator ex-
pression in Eq. ( I18p . The effect of longitudinally polarized beams on the polar averaged
rate (called total rate) can be obtained from the form
/
a = a(P cS = 0) (1 - h-h^
1 + P
eff
j _l_ 942 Hy+L \
941 On H u+L
9ll -, , #12 Hu+L
1 +
(27)
gu H u+L J
4 For the present purposes we neglect the 0(a s ) normal component of the polarization vector which
is quite small. Note that, in general, one needs two angles to describe the orientation of the polarization
vector instead of the one angle a defined in Eq. (|26
12
which, at the Born level and at y/s = 500 GeV, gives
a = a(P cS = 0) (1 - h-h+)(l - 0.37 P cS ) . (28)
From Eq. ( )28|) it is evident that the total rate becomes maximal on two counts: (i)
large values of the gain factor Kq = (1 — h-h + ), requiring sign(hJ) = —sign(h + ); and
(ii) large negative values of P e g, which can be achieved with large negative and positive
values of h_ and h + , respectively. The maximal enhancement of the rate will be obtained
for h- = —1 and h + = +1 such that P e g = —1 and Kq = 2. At y'l = 500 GeV, this
leads to a maximal enhancement factor of 2.74 over the unpolarized case. It is interesting
to note that for (bb) production at \/s = 500 GeV the effective enhancement through
beam polarization effects is slightly larger than in the (tt) case. For (bb) production the
last factor in Eq. (127)1 is replaced by the simpler expression (1 + P e g gu/gu) since, at
a/I = 500 GeV, the ratio Hf I+L /Hl r+L is practically zero for bottom quark production.
Using the results of the Appendix applied to the (bb) case, one finds ga/gu = —0.62
leading to an overall enhancement factor of 3.24 for the optimal choice of parameters
h- = -l and h+ = +1 (P cS = -1) at a/I = 500 GeV.
In Fig. [3b we show some contour lines for fixed values of the gain factor Kq =
(1 — h_h + ) in the /i + )-plane. Clearly, quadrants 2 and 4 are favoured if one wants
to obtain a gain factor exceeding one, i.e. Kg > 1. As concerns the rate dependence on
P e ff (rightmost factor in Eq. (EE}), a further rate enhancement is achieved for negative
values of P e ff , i.e. one would have to choose points lying to the left of the line h_ = —h +
in Fig. [3k. The optimal choice as concerns the rate would thus be quadrant 2 in the
(h-, /i + )-plane. One notes that large negative values of P e s can readily be achieved for
nonmaximal values of the beam polarization, as illustrated in Fig. |3k, where we have
plotted some contour lines in the (h_,h + ) plane corresponding to fixed values of P e g.
One notes that the regions of large Kg and large negative P e g have a large overlap. We
mention that one may have to give up the optimal choice in the (h_,h + ) plane if one
wants to achieve other goals such as minimizing the polarization.
The QCD one-loop corrections to the total cross section are well-known (see e.g.
Ref. [18J) and add about 13% at y/s = 500 GeV to the Born total cross section, where
the percentage increase has very little dependence on the beam polarization. We mention
that the electroweak corrections to the total rate are smaller, and amount to about 50% of
the QCD corrections [23]. For the strong coupling a s we use two-loop running adjusted
to the value a s (mz) = 0.1175 and fitted at 2m t = 350 GeV@ Close to threshold the
0(a s ) corrections become larger and amount to about 27% of the total cross section at
e.g. a/I = 400 GeV. The cm. energy dependence of the total cross section a is shown
in Figs. Hk and Hb. In Fig. H|l we take h + = and show the energy dependence of
5 For a we take the value a = 1/137. If one uses a running a, for example a = 1/128, the cross
sections in Fig. [4] would increase by 14.6%.
13
Figure 4: The total cross section a at the one-loop level as a function of the beam energy
yfs and (a) the electron polarization h_ (h + = 0); (b) for three values of the positron
polarization h + = 0, 0.4, 0.6, and with the fixed electron beam polarization of h- = —0.9
(solid lines). In Fig. Hb we also show the respective LO rates (dashed lines).
14
the total cross section varying h- over its whole range [— 1,+1]. One notes a strong
dependence on h_ apart from the standard falloff of the total cross section with beam
energy. Since for h + = the gain factor Kq is equal to 1 and since P e ff = h~, the
rate depends linearly on h- as displayed in Eqs. ( 1271) and ( )28l) . The rate is largest
for h_ = — 1 and then linearly drops to its lowest value at /i_ = +1. In Fig. 0b
we show the energy dependence of the rate for the three pairs of beam polarizations
(7i_, h+) = (-0.9, 0), (-0.9, +0.4), (-0.9, +0.6). If one translates this into the (K G , P eS )
representation, one has (Kq, P e ff) = (1, —0.9), (1.36, —0.96), (1.54, —0.97). The hierarchy
of rates in Fig. 0b can be seen to be mostly determined by the gain factor Kq in Eq. ( 1271) .
Next we turn to the differential rate distribution with respect to cos 9. In order to il-
lustrate the forward dominance of the differential cos ^-distribution we plot a -1 da j 'd cos 9
against cos 9. Note that the dependence on the gain factor Kq = 1 — h-h+ drops out
in the ratio. In Fig. we plot the differential rate distribution for a fixed value of
yfs = 500 GeV and for P e s = — 1,0, +1. One sees a pronounced forward dominance of
the differential distribution which does not depend much on the value of P e g. In Fig. [5b
we keep the effective beam polarization fixed at P e ff = and vary ^/i through several
values. At threshold y/s = 350 GeV one has a flat distribution a~ 1 d<j/dcos9 = 0.5.
When the energy is increased, the forward rate clearly dominates over the backward
rate. The forward dominance becomes even stronger for increasing energies.
Of related interest is the rate into the forward (F) and backward (B) hemispheres.
Again, the gain factor Kq drops out in the ratio. At y/s = 500 GeV, one numerically
obtains
The mean forward rate (<j)f clearly dominates over the mean backward rate (ct)b- The
dependence of the Fj B rate ratio on P c g is not very pronounced.
In Fig. |6] we plot the polar angle dependence of the NLO left-right polarization
asymmetry Alr for different energies. At y/s = 360 GeV the cos 9 dependence already
starts to deviate from the flat Born term behaviour at threshold given by = —((741 +
942) /(dii + 912) = 0.409. The left-right polarization asymmetry A LR peaks toward the
backward region and reaches ~ 59% at the backward point for the highest energy -y/i =
3000 GeV in Fig. El
4 Born term simplifications at threshold
and in the high-energy limit
Before turning to the numerical analysis of the polarization of the top quark, in this
section we shall first discuss Born term simplifications of the polarization of the top quark
(29)
15
Figure 5: Polar angle dependence of the differential cross section for (a) y/s = 500 GeV
and P e ff = — 1,0, +1 and (b) P c s = for beam energies y^s = 360 GeV (dotted line),
500 GeV (dashed line), 1000 GeV (dash-dotted line), and 3000 GeV (solid line)
16
<
70
60
50
40
30
1 1 1 1 1 1
1 1 1 1
Vs -
360 GeV "
Vs -
500 GeV "
Vs -
1000 GeV
vs —
juuu i^ev
1 1 1 1 1 1
i i 1 i i i i r
iii
-0.5
0.5
cos
Figure 6: Left-right polarization asymmetry A^r for sfs
(notation as in Fig. [5])
360, 500, 1000, and 3000 GeV
at threshold and in the high-energy limit. In Sec. 5 we discuss Born term simplifications
that occur for P e ff = =1=1.
At threshold v — > and in the high-energy limit v — > 1, the polarization expressions
become quite simple. At threshold, the polarization of the top quark is parallel to the
beam axis, regardless of the polar orientation of the top quark (see e.g. Ref. [21] )• In
fact, a large part of the beam polarization gets transferred to the polarization of the
top quark at threshold. For the Born term contributions the top quark polarization at
threshold can be ca
nominally given bjj^
culated from Eqs. (fT9l). (1201 and (see also Ref.
-Peff — Alr ,
P
n.,
It is
(30)
1 — P e sA LR
where Alr is the left-right beam polarization asymmetry (&lr — (Jrl)/{vlr + &rl) at
threshold (see Eq. (1251) ) and n e - is a unit vector pointing into the direction of the electron
momentum. In terms of the electroweak coupling parameters (see the Appendix), the
nominal polarization asymmetry at threshold y/s = 2m t is given by A^r = —(g^i +
942)/ {9u + 912) — 0.409. The simplification at threshold arises from the fact that, from
the four (L, S)v,a amplitudes (L, S)v,a = (0, l)v, (2, (1, 0)^, (1, 1)^ describing the
production of a spin-1/2 pair, only the S"-wave amplitude (0, l)y survives at threshold.
The suffices V and A denote vector current (V) and axial vector current (A) production.
Correspondingly, the combinations (g 4 i + g i2 ) and (gu + gi 2 ) contain only the vector
6 As discussed in Sec. 2, QCD binding effects significantly modify the naive threshold results in the
threshold region.
17
current coupling on the quark side.
The magnitude of the threshold polarization is given by
P c s — A LR
PceAlr
(31)
The threshold polarization is independent of cos6*, i.e. (\P\) = \P\. The polarization
vanishes for P cS = A LR independent of cos6>Q For P cff > A LR and P c g < A LR one
has P = \P\n e - and P = — \P\n e -, respectively, such that p(* r ) = +~|P|sin0 and
P® = ±\P\ cos 9. In particular, one has a 100% threshold polarization of the top quark
for P e g = ±1 with P = ±n e -.
Extrapolations away from P e g = ±1 are more stable for P c g = — 1 than for P e g = +1
as the slope of Eq. (}3~Tj) at P e g = ±1 shows. One has
d\P\ _ ± 1±A LR
dP cS 1tA lr
(32)
For P eS = —1 one has a slope of —(1 — A LR )/(1 + A LR ) = —0.42 while one has a much
larger positive slope of (1 + A LR )/(1 — A LR ) = +2.38 for P eff = +1. This substantiates
the statement made above and in Sec. 1 about the stability of extrapolations away from
P e fj = ±1. For example, keeping only the linear term in the Taylor expansion of Eq. ( l3~Tj) .
one has |P | = 0.98 for P cff = -0.95, while |P | drops to |P | = 0.88 for P cff = +0.95.
For energies above threshold the slope Eq. (1321) becomes energy and angle dependent.
We do not show plots of the slope at higher energies. We have, however, checked numer-
ically that the above statement about the stability of the |P | result at P e e = — 1 against
variations of P e g- remains true at higher energies in the whole angular range, where the
slope in the backward region has a tendency to be smaller than in the forward region.
As mentioned above, minimal polarization |P | = occurs for P e g- = A^ R = 0.409 for
all values of cos#. This again shows that an extrapolation away from P e g = — 1 is more
stable than an extrapolation from P e ff = +1 since one is much closer to the polarization
zero in the latter case. This observation will carry over to the P e ff- dependence at higher
energies.
In Fig. [7^ we show the threshold correlation of the angles a and 6 for different values
of P e ff. Starting at P e g = — 1 the two angles are related by a = 180° — 9 up to the
longitudinal polarization zero at P e ff = A^ R = 0.409 after which the correlation becomes
a = -6.
As the beam energy increases, the polarization vector of the top quark slowly turns
into the direction of its momentum (or opposite to it). Finally, in the high-energy
limit s — > oo, when v — > 1, the polarization of the top becomes purely longitudinal in
the helicity system such that |P| = |P^| since its transverse and normal components
7 Thrcshold simplifications for (qq) production have also been discussed in Ref. |26j . Similar simpli-
fications for polarization observables occur for the threshold production of gauge boson pairs |27j .
18
1.0
19
involve a spin flip amplitude and thus vanish as m t /\fs. Note that, although p(* r ) is
asymptotically suppressed, it is still sizable at y/s = 1000 GeV as Fig. [1] shows.
In fact, in the high-energy limit, one has P(cos#) = P^(cos#) • p t with
P«(cos0) =
(9u + 04i + Pcftjgii + 044)) (1 + cosfl) 2 + (g u - g 41 - P cS (gn - 044)) (1 - cosfl) 2
(gil + 044 + Pes(gi4 + 04l))(l + COS6 1 ) 2 + (011 - 044 - P e s(gi4 ~ 04l))(l ~ COS 9) 2
(33)
for the surviving longitudinal polarization. In the same limit, the electroweak coupling
coefficients take the numerical values gu = 0.601, g u = —0.131, g 4 i = —0.201, g 44 =
0.483, #i2 = 0.352, and #42 = —0.164. When P e s = —1 it is more convenient to switch
to the chiral electroweak coefficients fn/LR defined in the Appendix. One has (Jll =
-1.190; f LR = -0.434)
p(tif a\ ±-b L R .,, , ( IlrY (I ~ cos 6) 2
p( >(cose) = — with b LR = — — — . 34
1 + \Jll) (1 + cos6») 2
P (f) goes through zero for b LR = 1 which is solved by cos# = -{Jll - Jlr)/ {fn + Ilr)-
For P e g = +1 one has a similar simplification where the quantities on the right-hand
side of Eq. (g3D are replaced by b LR -> b RL and f LL / L R -» Irr/rl {/rr = -0.867;
f R L = —0.217). In this case P^ goes through zero for b R L = 1, or for cos 9 = —(f RR —
JriMUrr + fRL)-
At threshold the rate shows no cos# dependence since the (tt) pair is produced in a
S'-wave state. This is different in the high-energy limit when v — 1, where the forward
rate strongly dominates over the backward rate, as an inspection of the denominator of
Eq. ( |33|) shows. Since an accurate measurement of the polarization observables requires
large statistics, and thus large event samples, the issue of rates is an important one.
Numerically, one finds a(cos# = +1)/<t(cos0 = -1) = 9.23 (1 - 0.31P cff )/(l - 0.60P cff ).
The dependence on P e g is small. When averaging over the forward (P) and backward
(P) hemispheres, one finds {a) F /{a) B = 4.04 (1 — 0.31P o fj)/(l — 0.43P c ff), i.e. in the case
of unpolarized beams when P e ff = the rate in the forward hemisphere dominates over
the rate in the backward hemisphere by a factor of four with only slight dependence on
beam polarization. Comparing to Eq. fl29|) the forward dominance is more pronounced
in the high-energy limit than at \/s = 500 GeV.
Equation ( |33l) also very nicely shows how varying P e Q affects the longitudinal polariza-
tion . For the unpolarized beam case P e g = the longitudinal polarization P™' is neg-
ative (—31%) at the forward point (FP) cos# = +1 and positive (+60%) at the backward
point (BP) cos 9 = —1. For maximally polarized beams P cff = ±1, Eq. (|33|) can be seen
to satisfy the angular momentum conservation conditions, Eq. (|23|) . For P cff = ±1 the
longitudinal polarization monotonically increases/ decreases from the backward to the for-
ward point. It can be seen to go through zero at cos# = (gu — <?4i — 5 , i2+fi , 42)/(5 l i4— #44) =
20
-0.47 (~ 117.8°) for P cS = -1 and cos 9 = -(g n +g 41 -g 12 -3 42 )/(#i4+#44) = -0.60 ( ~
126.9°) for P eff = +1 (see discussion after Eq. (151]) ). Close to P e fj = ±1, the longitudinal
polarization zeros are only mildly dependent on P c g. There is a range of P c s values for
which the longitudinal polarization remains positive over the whole cos 9 range. This
is determined by the zeros of the coefficients of the angular factors in the numerator of
Eq. (133]) . The condition for positivity of reads
gU + #41 ^ D #14~ #41
; < Pes < • (35)
#11 + #44 #11 - #44
Numerically this translates into 0.31 < P e g < 0.60. The same bounding values determine
the vanishing of the polarization at the forward and backward points. At the forward
point, where the rate is highest, the polarization |P | can be made to vanish by setting
P e ff = —(#14 + #4i)/(#n + #44) = 0.31. At the backward point, one has zero longitudinal
polarization for P cS = (g u - g 41 )/ (g n - g u ) = 0.60.
All of this is illustrated in Fig. [7b showing the correlation between P e g- and the
angles a and 9. The steplike behaviour in Fig. [7b is associated with the vanishing of
the polarization at which points the polarization vector changes its direction by 180°.
At P fj = — 1 the polarization vector P is antiparallel to p t up to where P becomes zero
at 9 ~ 117.8°. From then on P is parallel to p t . Zero polarization and the location of
the step-like behaviour is slightly P e g- dependent and is shifted to lower values of a. For
0.31 < P e ff < 0.60 the polarization P is always parallel to p t . Finally, for P cff = +1
the polarization P starts off parallel to pt and turns antiparallel to pt after the zero at
cos 9 ~ 126.9°. Again the polarization zero and the associated step-like behaviour is
slightly shifted when one moves away from P e g- = +1.
Given the fact that the polarization turns from the beam direction to the momentum
direction (or its opposite) going from threshold to the high energy limit it would be
interesting to know how fast this transition occurs when the beam energy is ramped up
in the envisaged range of beam energies yfs ~ 2m t -j- 1000 GeV. In Fig. [8] we investigate
the energy dependence of the angle a for several values of P e fj for a scattering angle of
9 = 90°. In Fig. [8^l we consider three representative negative values of P e g. All three
curves start off with the threshold angle a = 90°. The growth of a does not depend
much on P e g but is still far away from the asymptotic value a = 180° at = 1000 GeV.
For positive values of P e g the dependence of a on P eS is more pronounced (see Fig. [8b).
For P eff = +1 and P e s = +0.5, one is getting closer to the asymptotic value of a = 0°
at y/s = 1000 GeV than for the negative values of P e g shown in Fig. [SH- The behaviour
of the P e fj = +0.25 curve differs from the two other curves since one has crossed a
longitudinal polarization zero between P eff = +0.5 and P c g = +0.25.
21
^ 1 50
CD
0)
s 130
120
1 10
100
90
I I | I I I ! I I I
-
-
I | I I I I | I I I I |
I I I I | I
i
-
P -
r eff
-1
----- Pe„ =
-0.5
Peff —
-0.25
l l I l l l l l l l
G =
l I l l l l I l l l l I
90°
i i i i I i
i
400 500 600 700 800 900 1000
Vs [GeV]
(a)
^100
i i i i i i i i i i i i i i i i i i i i i i i i i i i i i i
^ 80
2 60
+ 0.25
« 40
Peff —
+ 0.5
20
Pe« =
+ 1
=
90°
-20
-40
-60
-80
fi i i i i i i i i i
I I I I I I I I I I I I
i i i i I i
i
400 500 600 700 800 900 1000
Vs [GeV]
(b)
Figure 8: The top quark polarization angle a for a scattering angle of 6 = 90° as a
function of the beam energy for (a) negative values of P c g = —1,-0.5,-0.25 and (b)
positive values of P e s = +1, +0.5, +0.25
22
5 Born term simplifications for P e ff — Tl
As has been emphasized in the notable paper by Parke and Shadmi [5], the Born term
polarization formulas considerably simplify for the case of maximal effective beam polar-
ization P e fj = —1 which corresponds to a (e^,e^) configuration. Although designed for
the case of top-antitop spin-spin correlations, the results of Ref. [5] are easily adopted
to the case of single-spin polarization as also noted in Ref. [6]. From a practical point of
view the limiting case P e g = —1 is very interesting since, as was emphasized in Sec. 2,
one can get quite close to the maximal value P e fj = — 1 even if the beam polarizations are
not close to their maximal values. Similar simplifications occur for the case P e fj = +1. In
order to distinguish between the two cases we add the suffices LR and RL for quantities
derived for the case P c s = — 1 and P e fj = +1, respectively.
For the Born term case and in the limit P e g = —1, the polarized numerators ( jl~9l)
and ( 120]) take a factorized form,
A^(cosfl) = ~(f LL (cos6 + v)+f LR (cos6-v)\A LR (cose)2N c s, (36)
Aff(cos0) = lsmeVT^(f L L + fLR)A LR (cos6)2N c s, (37)
o
where the common factor Al R (cos 9) is given by
A LR (cos 9) = f LL (l + v cos 9) + f LR (l - v cos 9) . (38)
We have made use of the chiral electroweak coupling coefficients /ll/lr of Ref. [5] which
are simply related to our electroweak coupling factors (see the Appendix). One can
check that one can obtain N^ R (cos 9) in Eq. ( 1361) from the generic spin formula Eq. (1) of
Ref. [5] when one specifies to the helicity system with cos£ = +1. Similarly, one obtains
N^ R (cos9) in Eq. f l37|) when one specifies to the transversity system cos£ = 0. In each
of the two respective systems, one has to take the cross section difference cr(tt) —ff{t-l).
One can then determine the angle a enclosing the direction of the top quark and its
polarization vector by taking the ratio /N^\ One has
r(*r)
_ jyW(oosg) sm9VT—^(f LL + f LR )
iVS(cos0) f LL (cos9 + v) + f LR (cos9-v)- ^
For example, at threshold (v = 0) one has tana^jj = — tang with the solution cxlr —
180° — 9 in agreement with the corresponding limit in Sec. 4. As another example we take
9 = 90° and obtain tana LR = -{f LL + f LR )/{f LL - f LR ) ■ VT^/v. For ^ = 500 GeV
this gives aL R = 124.9°, i.e. the polarization vector is still close to its threshold value of
&lr = 90° but has started to turn to its asymptotic value of am = 180°.
Equation (1391) is nothing but the defining equation for the off-diagonal basis in Ref. [5]
considering the fact that their angle £ is related to olr by £ = 180° — a^ R . In the
23
coordinate system where the z axis is defined by the angle a^R given in Eq. (139|) . the
polarization vector of the top quark is purely longitudinal. In particular, this means
that its transverse component is zero in the off-diagonal basis implying that the density
matrix of the top quark is diagonal in this basis. In this sense the "off-diagonal" basis
is a diagonal basis and the wording "off-diagonal" used in Ref. j5] for this basis can lead
to a misunderstanding.
A different but equivalent view on the off-diagonal basis may be obtained by rotating
the nondiagonal helicity system density matrix of the top quark (m,n = ±1/2)
l(a-l + i-a). (40)
in the scattering plane by an angle a. One has
' - A l l 2 ( \ jV 2 U *\
Pm'n' — a m'm\ a ) PmnO-nn' \ a )
1 + (PW cos a + P^ sin a) pW s i n a _ pM cos a
sin a - P(* r ) cos a 1 - (PW cos a + P^ sin a)
jl/2
2
where d^^a) is the usual spin-1/2 Wigner rotation matrix and |P | = V PW 2 + PW
It is evident that a rotation by the angle a = o>lr defined in Eq. fl39|) diagonalizes the
original density matrix as indicated in the last line of Eq. f Hdj) .
The correlation between the angles a (for general values of P e g) and 9 are shown in
the contour plots Fig. [9j In Fig. |9k we choose P e s = — 1 and show fixed energy contours
in the (a, 6*)-plane for several values of the cm. energy y/s. Up to y/s = 1000 GeV the
correlations do not deviate very much from the threshold correlation a = 180° — 9. In
the limit v = 1 one has a steplike behaviour of the correlation function as discussed
before in Sec. 4. In Fig. [9b we show the same plots for P c g = +1. The approach of the
correlation curves to the steplike behaviour at v — 1 is somewhat faster than in the case
P e fj = —1. In Fig. [9fc we show the same curves for P e g = 0.5 where one is close to the
polarization zero. The y/s = 360 GeV correlation curve is still close to the corresponding
threshold curve a = —9. At higher energies one sees a different behaviour in as much as
the correlation curves run into a = 0° at the backward point as does the flat asymptotic
curve as discussed before in Sec. 4 (Fig. [7]).
In order to calculate the normalized polarization components one needs also the
denominator factor D(cos9) in Eq. (122]) . again for the Born term case and P e g = — 1.
One has
D LR (cos 9) = ^ (A\ R - 2f LL f LR v 2 sin 2 9) 2N c s . (42)
proportional to the cross section sum a(tt) + ff{t\) in any of the systems in Ref. [5].
24
to
CD
CD
cn
(D
150
00
50
Vs = 360 GeV
Vs = 500 GeV
Vs = 1 000 GeV
Vs = 3000 GeV
50
100
1 50
[degrees]
(a)
CD
CD
CP
CD
X3
-50
-1 00
-150
Vs = 360 GeV
•— Vs = 500 GeV
---- Vs = 1 000 GeV
Vs = 3000 GeV
Paf. = +1
J I I I I L
50
100
1 50
[degrees]
(b)
o
CO
CD
CD
CD
CD
TO
-50
00
50
Vs = 360 GeV
Vs = 500 GeV
Vs = 1 000 GeV
Vs = 3000 GeV
50
100
1 50
[degrees]
(c)
Figure 9: Correlation of angles a and 9 for (a) P eS = —1 (a = oilr), (b) P c s — +1
(a = am,), and (c) P c s = +0.5 for different values of the cm. energy yfs = 360, 500,
1000, and 3000 GeV (notation as in Fig. EJ
25
Using Eqs. ( 155|) and (|4"2]) the longitudinal polarization P^ = Nf R /D LR can be seen
to become maximally —1 and +1 in the forward and backward directions, respectively,
in agreement with angular momentum conservation as before. One also reproduces the
threshold formula Eq. f )30|) and the high-energy formula Eq. ( 133]) when these are specified
to P c ff = — 1. The longitudinal polarization goes through zero at
cos0 o = — — v = v (=— 0.48-u). (43)
ILL + JLR 011 — 041 + 012 — 042
At this value of cos 9 the polarization vector of the top quark is orthogonal to its mo-
mentum. Later on we shall see that, at this point, p( tr ) acquires its maximal value and
|P | acquires its minimal value. Since the ratio (fn — Ilr) / {III + Ilr) is only mildly
energy- dependent, the location of the zero is mainly determined by the velocity of the
top quark, i.e. it moves towards the backward point when the energy is increased. For
convenience we have added the y/s = 500 GeV value of the electroweak coupling ratio in
brackets in Eq. fj43l .
The transverse polarization Pf R vanishes in the forward and backward directions
due to angular momentum conservation, as is explicit in Eq. ( 157)) . It becomes maximal
at the point where the longitudinal polarization goes through zero. This can be verified
by an explicit calculation, viz.
4^4 = 0, (44)
dcosO cqs9q
where cos 9q is given in Eq. (143]) .
Whereas there are no illuminating expressions for the longitudinal and transverse
polarization components for general values of the velocity v, the magnitude of the po-
larization |P | for P e ff = — 1 takes the simple form
Li? I
V -l-2a 2 LR -8al R -18al R ..., (45)
Dlr 1 - 2a LjR
where the coefficient cllr depends on cos 9 through
aLR{cose)= AuiZf 2 ^ 6 - (46)
The convergence of the expansion in Eq. fj4"5"]) is rather slow except for very small values of
cllr. Note that the expansion in Eq. fj4"5"]) deviates from 1 only at 0(a 2 LR ). At the forward
and backward points where am = 0, one has \Plr\ = 1 as stated before. Between the
forward and backward points the polarization remains reasonably large. For example,
for \fs = 500 GeV the polarization never drops below |Pl_r| = 0.95. Differentiating of
Eq. (145!) with respect to cos 8 one can see that the minimum of \Plr\ occurs at the
point where the longitudinal polarization pf R vanishes (see Eq. ( 1431) ). i.e. where the
polarization is purely transverse. The high-energy limit of Eq. ( 1451) is discussed in Sec. 4.
26
Figure 10: Difference Aa iR = a£#(NLO) — o>lr(IjO) of NLO and LO polarization angles
for v/i = 360, 500, 1000, and 3000 GeV (notation as in Fig. EJ
Similar simplifications occur for the case P e fr = +1 which corresponds to the (e^, ej)
configuration. This is effected by the replacement /ll — > /rr and /lr — > Jrl with a
corresponding change in the notation A LRj cilr — > A rl ,clrl- Further one has Nr[ =
-Nf R {f L L -> Jrr-Jlr Jrl) and ivgg = -JVg } (/ LL /rh; /la -> /rl). The
zero of P( £ ) is now located at cos#o — —0.63?; for y/s = 500 GeV, i.e. the zero is closer
to the backward point than in the case P c s = —1. For 9 = 90° the angle am can be
calculated from tan a RL = rr~\~ f rl) / (Irr ■'n/I - v 2 /v which, at y/s = 500 GeV,
gives ctRL = —48°. At cos6> = and y/s = 500 GeV, one has a LR > a RL leading to
\Plr\ < \Prl\-, i-e. the (e^, ej) configuration leads to larger values of the polarization
than the (e£,e^) configuration at this point of parameter space. In Fig. [9b we show a
contour plot in the (curl, 0) plane for several values of the cm. energy y/s. As is the case
for the (aiR, 0) correlations, the (cxrl, 0) correlations do not deviate very much from the
threshold correlation a = —9 up to y/s = 1000 GeV.
As a last point we discuss how the polarization angle a. j_, R changes when going
from LO to NLO. In Fig. [10] we show a plot of the cos# dependence of the difference
ActLR = «£r(NLO) — «lr(LO) for different energies. The maximal values of the differ-
ence occur at values of cos# where the polarization vector is perpendicular to the top
quark's momentum, i.e. where cx,lr = 90° (see discussion after Eq. f !43p ). The difference
can become as big as 10° for y/s = 3000 GeV. The radiative corrections can thus be
seen to rotate the polarization vector away from the off-diagonal basis by a nonnegligible
amount.
27
6 The polarization components p( tr \ and P^
We now turn to the numerical discussion of the three polarization components
p( tr \ and keeping in mind the Born term simplifications for P^ and p( tr ) discussed
in Sees. 4 and 5. We start our discussion with the longitudinal component P^K In
Fig. ITTa we show the dependence of the NLO longitudinal polarization P^' on cos 9 at
y/s = 500 GeV for several values of P e ff spanning the whole parameter range of P e g.
The dependence of P^ on P e ff and cos9 is quite pronounced. For P e g = ±1 the cos^-
dependence already deviates considerably from the (Born term) threshold behaviour
cos#. It is quite interesting to observe that all NLO curves intersect at one
point where cos9 = —0.406. This can be verified by setting to zero the derivative of
p( £ ) with respect to P e s. The relevant higher order equation admits of a solution at the
above value of cos 9. In Fig. [TTb we show the cos ^-dependence of P^ for several energies
keeping P e g- fixed at P e g = — 1. At the resolution of the figure all curves seemingly go
through —1 and +1 at the forward and backward point, respectively, showing that hard
gluon emission effects are not very strong at these energies. The energy dependence is not
very pronounced, even if the y/s = 500 GeV curve already deviates from the threshold
behaviour P™> = — cos 9.
The strong dependence of on P e g can be nicely exposed by considering the LO
expression for the polar mean (P^) which is obtained by integrating the numerator and
the denominator in Eq. ffl8|) separately over cos 9. One obtains
/pM\ = <^!> = t v 9_u + ^Pff m
V 1 (D) 3 (gii+g 4 iPes)(l + v 2 /3) + (g 12 + 9 4 2Pcs)(l-v 2 ) ' V ;
(pW) vanishes at threshold. In the high-energy limit, one has (P^) = (<?i4+g44P e ff) / (gn +
c/4iP e ff) which, for P cS = ±1, gives (P^) = 0.882 and (P^) = -0.766 close to the
y/s = 1000 GeV values in Fig. [TJ At y/s = 500 GeV, one has
<P W > = (P W )(P cff = 0) \_ 3 Q f 7 p2 = { -0.15 = 0\. {41
One observes a strong dependence of the mean longitudinal polarization on P e g. By
comparing with the y/s = 500 GeV point in Fig. [T^i, one observes a 2% change in (P^)
due to the radiative corrections.
The same strong dependence on P e g- is found when one averages over the forward
hemisphere where one has (y/s = 500 GeV)
{P m )F = {P (i) )F (Peff = o) \_ 3 Q 2p* = { -0-27 = }. (49)
+0.62
Peff = +1
-0.15
=
-0.50
= -1
+0.85
Peff = +1
-0.27
=
-0.81
= -1
28
00
00
50
■50
00
Vs = 500 GeV p eff = -1
_J I I I L_
_J I I L_
-0.5
0.5
cos
(a)
=v I I I I I I I
I I I I
Vs =
III
360 GeV I
Vs =
500 GeV -
Vs =
1000 GeV 1
Vs =
3000 GeV
p -
r eff
-1
I I I I I I I
II 1 1 -
-0.5
0.5
cos
(b)
Figure 11: NLO longitudinal top polarization as a function of cos 9 drawn (a) for effective
beam polarizations P e ff = — 1, —0.5, 0, +0.5, +1 (notation as in Fig. [TJ at y/s = 500 GeV;
(b) at beam energies y/s = 360, 500, 1000, and 3000 GeV (notation as in Fig. EJ) and
Peff = "I
29
^ 1 00
A
o_ 50
v
-50
100
I I I I I I I I I I
I I I I I I I I I I I I I I I I I I
forward
backward
P^=-1 "
I I I I I I I I I I I I I I I I I I I I I I I I I I I I I
400 500 600 700 800 900 1000
Vs [GeV]
(a)
A
V
1 00
80
60
40
20
I I I I I I I I I I
1 1 1 1 1 1 1 1 1 1 1
forward
backward -
1 I 1 1 1 1 I I I I
I 1 1 1 1 I 1 I I I 1 1 1 1 I 1 1 1
400 500 600 700 800 900 1000
Vs [GeV]
(b)
Figure 12: Average (a) longitudinal polarization (P^) and (b) transverse polarization
(p(* r )) in the forward and backward hemispheres for P e fr = — 1
30
When one averages over the backward hemisphere the average longitudinal polarization
is smaller and the dependence on P e g- is much weaker viz.
, Peff = +1
(P») B =(P<"> B (^ = 0)i^i^=| + 0.13 = _0
(50)
In Fig. [T2k we show a plot of the energy dependence of the forward and backward
averages of the longitudinal polarizations for P e fr — — 1- The forward average
is large and negative. It starts with a nominal threshold value of (P^')f — —0.5 and
slowly drops to a value of (P^)f — —0.90 at \/s = 1000 GeV which is not far from
the asymptotic Born term value (P {e) ) F = ~{7f 2 LL - f 2 LR )/(7f 2 LL + f 2 LR ) = -0.96. The
backward average b is smaller and positive. It drops from a nominal threshold
value of (P W ) B = +0.5 to (P W ) B = +0.14 at ^ = 1000 GeV as compared to the
asymptotic Born term value (P®) B = ~(f 2 LL - 7f 2 LR )/(f 2 LL + 7f 2 LR ) = -0.04.
We now turn to the transverse polarization component P^ tr \ Similar to Figs. [TTb
andfTTb we show the corresponding curves for p(' r ) in Figs. [T3k and [13b. The transverse
polarization vanishes at the end points due to the overall sin 9 factor in the angular decay
distribution Eq. (1201 . The dependence on P e g is again quite pronounced. One observes a
faster change with P e g- at P e e = +1 than at P e s = — 1. p(' r ) vanishes close to P e ff = +0.5.
At y/s = 500 GeV the deviations from the threshold behaviour p(' r ) = ±\/I — cos 9 2
for P c g = ±1 are slight but clearly visible. The (v^) _1 dependence of the transverse
polarization is easily discernible in Fig. [T3b . A visual inspection shows that, as is exact
in the Born term case for P e g = ±1, the NLO longitudinal and transverse polarization
components are complementary in the sense that the transverse polarization becomes
maximal very close to the point where the longitudinal polarization becomes minimal,
and vice versa. This observation bodes well for the existence of large values of the total
polarization as discussed in Sec. 7.
For the mean value of the transverse polarization, one obtains the Born term level
expression
/p(tr)\ = n™* 041 + 942 + (gll + ffl2)Pcff ^
{ '~ 2^( gil +g 41 P eS )(l + v 2 /3) + (g 12 + g 42 P cS )(l-v 2 )' 1 '
At nominal threshold one has (p(* r )) = =p7r/4 for P e fj = ±1 close to what is seen in
Fig. [TJ In the high energy limit (P^) vanishes as (y/s) _1 . At y/H = 500 GeV one has
f -0.57 P cff = +1 ]
{ p(tr) } = {P (tr) ){PeS = o) [I 2 ™*" = +0.24 = , (52)
"°' '"" [ +0.61 = -1 J
showing again the large effect of beam polarization. By comparing with the \fs =
500 GeV point in Fig. [lb, one observes a 1.5% change in (p(* r )} due to the radiative
corrections.
31
Figure 13: NLO transverse top polarization as a function of the scattering angle 9 drawn
(a) for effective beam polarizations P e ff = — 1, —0.5, 0, +0.5, +1 (notation as in Fig. [T|) at
= 500 GeV; (b) at beam energies = 360, 500, 1000, and 3000 GeV (notation as
in Fig. [5]) and P eff = — 1
32
For the ratio of the forward and backward mean of the transverse polarization, one
obtains
(P^) F (N^) F (a)
B
(PM) B (M»)) B (a),
+0.54
P cS = +1
+0.53
=
+0.61
= -1
(53)
There is a slight dominance of the backward mean as also evident in Fig. [13j The
dependence of the ratio (153]) on P eff is not very pronounced.
A plot of the energy dependence of the forward and backward averages of the trans-
verse polarization is shown in Fig. IT2b . Both curves start at the nominal threshold value
(P^O)^ = (p( tr )} B = 71-/4 and then quite slowly begin their descent to their asymp-
totic demise. At y/s = 500 GeV and P e g = —1, one can compare the NLO result for
(P {tr) ) F /(P {tr) ) B = 0.68 with the corresponding LO result (P (tr) ) F / (P (tr) ) B = 0.61 in
Eq. §3D.
The normal polarization component P^ is a T-odd observable and thus obtains
only contributions from the imaginary parts of the production amplitudes. Since we
are neglecting the contribution from the imaginary part of the Z propagator the only
contribution to the normal polarization component P^ at 0(a s ) is that of the imaginary
part of the one-loop contributions (Eqs. (fT4"l) and (115jl ). When averaging over cos#, the
contributions of Hj' 2 ^ n \loop) drop out and one has the 0(a s ) result
[ } S 6v/i l ) (9n + 94iP eS )(l + v 2 /S) + (g 12 + g 42 P eS )(l-v')- 1 }
Numerically one has (y/s = 500 GeV; a s = 0.094)
(P (n)) = (pW)(P eff = ) * ~ = { -0,013 = 0\. (55)
Clearly the normal polarization component is small being an 0(a s ) effect. Also, the
dependence of (P^) on the beam polarization is quite small.
In Fig. [IH we show the cos 9 dependence of the normal component of the polarization
of the top quark. In Fig. H4"a we keep the energy fixed at y/s = 500 GeV and vary P e g.
The differential distribution peaks at around cos 8 = where the peak moves to the left
with increasing values of P e g . The peak values of P^ 1 are around —2%. The dependence
on P e g is weak as also evident in Eq. ( |55|) . In Fig. [14b we plot the cos 9 dependence for
different energies keeping P e g fixed at P e g = — 1. As expected, the normal polarization
can be seen to decrease with the typical (i/i) _1 behaviour. We mention that we are now
in agreement with the results of Ref. [1] when one takes account of the fact that their
normal direction is defined opposite to ours.
Let us close this section by comparing our results to those of the authors of Ref. [6]
who calculated 0(a s ) radiative corrections to rates into definite spin states in generic
0.015
PcS = +1
0.013
=
0.012
= -1
33
Figure 14: 0(a s ) normal polarization of the top quark as a function of cos# (a) for
effective beam polarizations P e s = —1, —0.5, 0, +0.5, +1 (notation as in Fig. CD) at y/s =
500 GeV; (b) for P cS = -1 at beam energies ^ = 360, 500, 1000, and y/s = 3000 GeV
(notation as in Fig. [5])
34
coordinate systems starting from the initial beam configurations (e^,e^) and (e^,e£)
which correspond to P c g = — 1 and P e ff = +1, respectively. Put in a different language,
they compute radiative corrections to the (unnormalized) diagonal spin density matrix
elements cr-j- and crj,. In the helicity system, where they use the notation = (?l/r,
their polarized rate ol/r are related to our longitudinal polarization component P^' via
0l/r = \(1*P®) (56)
Comparing to the (e£,e^) subdominant spin rate ratio at = 400 GeV in Table 3 of
Ref. [6] we find a —1.34% reduction relative to the LO rate ratio vs. their reduction of
— 1.19%. We consider the two results to be consistent with each other within rounding
errors. We mention that our NLO results have been checked before in Ref. jl].
7 Total polarization and
orientation of the polarization vector
The magnitude of the polarization (also called total polarization) is given by
|P | = ^(PW) 2 + (P( fr )) 2 + (PW) 2 . (57)
In Fig. [TBI we show the NLO dependence of |P| on cos 9 at different values of y/s for
the three different values of P e s = ±1 and 0. As a general feature one observes that the
magnitude of the polarization decreases with energy. When P e fj = ±1 one obtains large
values of |P |, in particular in the forward hemisphere. For example, for yfs = 500 GeV
\P | remains above 95% over the whole angular range for P e fj- = ±1. The polarization
is slightly larger for P c q = +1 than for P e s — — 1. At y/s = 360 GeV and P e ff = ±1
one is still very close to the flat threshold behaviour |P | = 1, whereas for P e g = one
observes a slight falloff behaviour going from the backward to the forward point. Even
for the largest energy yfs = 3000 GeV, one does not have a zero for |P | showing that one
is still away from the asymptotic v — 1 case since asymptotically one has polarization
zeros for the three cases P e s = ±1,0 as discussed in Sec. 4 and exhibited in Fig. [7J As
mentioned before there is also a very small 0(a s ) normal component of the polarization
vector which will contribute to |P | at the 0(0.01). It is so small that it is not discernible
in our numerical plots.
In Fig. dni we show a plot of the energy dependence of the polar average (|P|)
of the total polarization. For both P e fj = ±1 the average polarization is large in the
whole energy range with a slight decrease when the energy is increased. The P e g =
+1 polarization is slightly larger than the P c g = —1 polarization. The average total
polarization becomes smaller when the effective polarization is reduced from P e fj = ±1.
As has been discussed before the rate of decrease is much faster for P e fj = +1 than for
35
o
-1 -0.5 0.5 1
cos
-0.5 0.5 1
cos
(c)
Figure 15: Total NLO top quark polarization as a function of cos 9 for beam energies
yfe = 360, 500, 1000, and 3000 GeV (notation as in Fig. ED and (a) P cff = -1, (b)
P eff = +1, and (c) P cS =
36
^ 80
60
40
20
1 T — ]■ -f-^-i-i. 4 _ ^_ J _ 1 _
1 " " 1 1 '
_ef f _
1 1
+ 1.0 -
P«ff =
-1.0-
: o = 90°
' eff =
-0.5 -
Pe« =
0.0 ~
i i i i i i i i i
iii
Peff =
1 1 1 1 1 1 1 ill
+ 0.5 "
i i
400 500 600 700 800 900 1000
Vs [GeV]
Figure 16: Average NLO top quark polarization (\P\) for a scattering angle of 9 = 90° as
a function of the beam energy y/s for P e g = — 1, —0.5, 0, +0.5, +1 (notation as in Fig. [TJ)
P e ff = —1 as can be appreciated by comparing the P e g = —0.5 and P e g = +0.5 curves.
The smallest polarization in Fig. [16] is obtained for P e g = +0.5. As will be discussed
further on P c g = +0.5 is close to the effective beam polarization where one has minimal
polarization.
Of interest is the total polarization in the forward and backward hemispheres. In
Fig. El we show plots of the average total polarization ( |P | )f/b for P c g = ±1, 0, where
the averaging is done over the forward and backward hemispheres. The average total
polarization ( |P| )p in the forward hemisphere is quite large for both P c g = ±1 and
remains larger than 95% even up to y/s = 1000 GeV. This is quite welcome from the point
of view of statistics since the bulk of the rate is in the forward hemisphere. The P c g = +1
polarization is slightly larger than the P e g = — 1 polarization. The average backward
polarization (\P\)b is significantly smaller than the forward polarization (\P\)p for
both P cff = ±1 as can also be appreciated by looking at Fig. [151 Both forward and
backward P e g = polarizations show a slightly decreasing energy behaviour starting at
the common threshold value of ( |P | ) = Arl — 0.409.
Returning to Fig. [T5b (P e g = +1), one observes a conspicuously large 10% radiative
correction at the backward point for y/s = 3000 GeV where the Born term prediction is
|P | = 1. One can attempt to understand this large value by substituting the asymptotic
values of the radiative corrections calculated in Ref. [28]. For the surviving longitudinal
37
- 100
A
to_
V
80
60
40
20 -
Peff = " 1
Peff=
Peff = + 1
J I I I I L
J I I I I I L
_L
F 1
I I I I I I I L
400 500 600 700 800 900 1000
Vs [GeVJ
Figure 17: Total NLO top quark polarization averaged over the forward and backward
hemispheres for P e g = —1, 0, +1 function of \/s
component one obtains
P {1) = -(l-|(|P + [2]) + -..), (58)
where the bracketed notation "[2]" denotes the anomalous contribution not present in
mtjsfs = production (see Ref. [28]). Using f 2 RR / f 2 RL = 16.069 and a s (3000GeV) =
0.079 the radiative correction at the backward point amounts to 15% which is reason-
ably close to the value in Fig. [15b. The anomalous contribution is quite small. The
corresponding formula for Fig. [T5k (P cff = —1) is obtained from Eq. fl5"g|) by the substi-
tution f R R,RL —> Ill,lr- With j\ h = 1.417 and f\ R = 0.188, one obtains a radiative
correction of 8% at the backward point, again in approximate agreement with Fig. [T5r .
One may state that the large radiative corrections at the backward point for P e g- = ±1
at \fs = 3000 GeV result from the fact that f RR ^> f R i and fn 3> Ilr-
Next we investigate the parameter space for which the polarization of the top quark
is minimal. For some measurements it may be advantageous to eliminate or minimize
polarization effects. For once, one can thereby gauge the efficiency of a polarization
measurement against an unpolarized control sample. The parameters to be varied are the
effective beam polarization P e g, the polar angle 8, and the energy ^/i. The minimization
is done at NLO including the normal polarization component according to Eq. ( !57|) .
In Fig. [T8k we show a plot of the NLO values of P e ff which minimize |P| for any
given scattering angle. The minimizing values P c ™ n depend in addition on the energy.
An important feature of the minimizing effective beam polarization is that, in the for-
ward region, where the rate is largest, the dependence of P c ™ n on cos 8 is reasonably
38
E 0.8
E ?
CL
0.6
0.4
0.2
-1 -0.5 0.5 1
cos
(a)
^ 20
IK
15
10
5
-1 -0.5 0.5 1
cos
(b)
Figure 18: (a) P e g values required for minimal top quark polarization \P | and (b) minimal
values for \P\, plotted against cos8, for y/s = 360, 500, 1000, and 3000 GeV (notation
as in Fig. [5])
39
flat for all shown energies. This means that it is possible to tune the effective beam
polarization in the forward region for each energy such that one obtains approximate
minimal polarization. Just above threshold at y/s = 360 GeV, P c g in is close to the flat
behaviour at threshold P^ m = Arl = 0.409. Apart from the near-threshold curve P c r ff m
shows a strong dependence on cos 9 in the backward region. The corresponding minimal
values of \P | are shown in Fig. [18b. At the forward and backward point the minimal
polarization is zero by construction. In the forward region the polarization remains quite
small starting from zero at the forward point. This is different in the backward region
where the minimal polarization can become as large as 18% for the highest shown energy
of y/s = 3000 GeV.
We now turn to the orientation of the polarization vector. We have already discussed
some aspects of the orientation of the polarization vector of the top quark in Sees. 4
and 5. We now combine the information on the orientation and the magnitude of the
polarization vector in one single (radius, angle) plot where we trace the end point (apex)
of the polarization vector P within the unit circle while increasing the energy from
threshold to infinity. The apex stays within the unit circle since \P\ < 1. In Fig. [T9~r
we consider the case P e s = —1 for the polar angles 9 = 60°, 90°, 120°, and 150°. All
trajectories start off at threshold where \P | = 1 and a = 180° — 9 and, depending on
cos#, end up at a = 0° or a = 180° with a length close to the asymptotic Born term
result (Eq. ( 134|) ). Which of the two asymptotic solutions a = 0° and a = 180° are
attained can be traced to the corresponding LO result (Eq. ( )34l) ) or from Fig. [7b- For
P e fT = +1 (Fig. IT9b) the trajectories start off at threshold with a = —9 and end up
at a = 0° or a = 180°. The appropriate solution can again be read off from the Born
term formula (Eq. fl34|) ) with the appropriate replacements as described after Eq. f )34|) .
or from Fig. [7b- The length of the asymptotic polarization vector is close to what is
obtained from Eq. f l34"l) after the appropriate replacements. Since f\ R j ' f1 L > f\ L j ' f RR
the asymptotic values of | P | and thereby the intermediate values of | P | are larger for
P e fj = +1 than for P eS = — 1. We remind the reader, though, that extrapolations away
from P fj = —1 are more stable than extrapolations away from P e s = +1. This is
illustrated in Figs. fl~9a and IT9b by adding the corresponding trajectories (dashed lines)
for P c g = —0.95 and P c g = +0.95, respectively. One observes only a minor change in
Fig. [T9k going from P e g = — 1 to P c ff = —0.95. For example, the total polarization |P |
remains close to maximal at y/s = 500 GeV for the technically feasible effective beam
polarization of P e ff = —0.95. The corresponding changes in Fig. fl~9b are much larger.
In particular, the total polarization |P | at P c s = +0.95 is considerably reduced from its
values at P e fj = +1.
In Figs. [T9b and [19b we have marked the energy dependence of the polarization vector
by dots (or ticks) on the trajectory of the apex of the polarization vector. One notes that
there is very little change in the length of the polarization vector going from threshold to
40
Figure 19: NLO Parametric plot of the orientation and the length of the polarization
vector in dependence on the cm. energy y/s for values 9 = 60°, 90°, 120°, and 150° for
(a) PcflF = —1 (solid lines) and P cff = —0.95 (dashed lines), and (b) P e fr = +1 (solid lines)
and P cS = +0.95 (dashed lines). The dots on the trajectories from the border to the
central line stand for ^ = 500, 1000, and 3000 GeV
41
\fs = 500 GeV. The ticks are approximately equally spaced on the trajectories indicating
an approximate inverse power law dependence of the spacing on the energy. For the three
trajectories 9 = 60°, 90° and 120°, the angle a is monotonically increasing with energy.
In contrast to this the 9 = 150° trajectory shows a kink at around \fs = 500 GeV. Both
Figs. H~9a and [T9b show that at \fs = 3000 GeV one has not yet reached the asymptotic
regime.
8 Longitudinal spin— spin correlations
Up to this point we have only discussed the single-spin polarization of the top quark.
The polarizations of pair produced top and antitop quarks are correlated and could
be observed in the energy spectra of decay products, especially in the energy spectra
of leptons and antileptons. There are altogether nine double-density matrix elements
describing the spin-spin correlations of the top and antitop quarks. Here we concentrate
on the longitudinal spin-spin correlation which is the double-density matrix element that
survives in the high-energy limit (for analytical NLO results see Refs. [281 EHJ |30]). We
mention that the full set of NLO double- density matrix elements has been numerically
evaluated in Refs. [31] 132] .
The longitudinal spin-spin correlation cross section is defined by
a.
(fi«0 = a Q (tt) - <r«(U) ~ Mlt) + MU) , (59)
where e.g. (tt) denotes a top quark with helicity 1/2 and an antitop quark with helicity
1/2, etc. Similar to Eq. the differential cos 9 distribution is given by
' l(7[!,i ' 2] 3 ;i + cos 2 9) of 1 * + \ sin 2 9 af^ + 3 cos 9 of 1 * , (60)
dcos# 8 V ' " 4 ^4
where
4t ] = (1 - h-h + f-0 (( 9 n + P«g«) + ( 9l2 + P cS942 )
4 lh) = {l-h-h+^fa + Prtg^H*™. (61)
The Born term contributions read [28]
Hi} tlta \Born) = -2iV c g 2 (l + v 2 ) } H 1 ^ 1 **' \Born) = N c q 2 (l - v 2 ) = H 2 L {lll2) (Born),
Hi {ei£2 \Born) = -2N c q 2 (l - v 2 ), H F {ilh) (Born) = -4N c q 2 v. (62)
Note that one has the Born term relations
H]j 2 (Bom) = -Hij 2{eie2) (Born),
Hp(Born) = -H F {ili2) (Born),
Hl' 2 (Born) = H^ 2 ^ (Born) , (63)
42
which are due to angular momentum conservation in the back-to-back configuration of
the Born term production |28j. These relations no longer hold true in the case of ad-
ditional gluon emission. The relations (J63l) imply that p(^ 2 ) = —1 at cos 6* = ±1
independent of P e g. Since the transverse contributions Hu,f dominate over the longi-
tudinal contributions one anticipates from the relations f )63|) that the longitudinal
spin-spin correlations are negative and only weakly beam polarization dependent.
Similar to Eq. f fl8|) . the cos# dependent longitudinal spin-spin correlation is defined
by the ratio
with the denominator function given in Eq. (j22p . The numerator function is given by
JV<**)(cos0) = |(1 + cos 2 £) {{g n + g^H 1 /^ + (g 12 + g i2 P cS )H 2 /^
+ - sin 2 ((^n + g 41 P cS )H l L (tlt2 > + (<? 12 + g 42 P cS )H z L
+^cos9 (gu + QMHf 1 ^. (65)
Let us first consider the polar angle average of the longitudinal spin-spin correlation.
For the Born term contribution, one obtains
1 (5n + feff<?4i)(l + 3w 2 ) + (g 12 + P c fr#42)(l
2\
^ ' 3 (pil + Peff^4l)(l+^ 2 /3) + (^12 + Peff^42)(l-^ 2 ) '
Note that = -1/3 at threshold (v = 0) and (p(^)) = _i i n the high-energy
limit (f = 1) independent of the beam polarization parameter P e fj. In fact, the depen-
dence on P e ff is very weak also for energies intermediate between these two limits. For
example, for y/s = 500 GeV one finds
(P (W 2 )\ = (p(^)up = 0) 1 °- 36Pcff
V 1 V /v ott ; i_o.37P cff
0.67
Pcff = +1 '
0.65
= >
0.65
= -1
• (67)
Equation (IBTj) shows that the dependence on the beam polarization parameter P e g prac-
tically drops out in the ratio f l67|) .
In Fig. [20] we plot the average longitudinal spin-spin correlation function (p(^ 2 )} up
to 0(a s ) as a function of y/s for different effective beam polarizations. The dependence
of (p(^ 2 )} on P e ff is barely visible.
As shown in Ref. [5] the off-diagonal basis defined by Eq. (|39|) diagonalizes both the
single-spin and the spin-spin double-density matrix at the Born term level for P e g = — 1.
In the NLO calculation described in this section the top and antitop quark are no longer
back to back due to hard gluon emission, i.e. in the two helicity basis (top and antitop
quark) the two respective z axis are not in general back to back. In the high-energy
43
Figure 20: NLO beam energy dependence of the polar average of the longitudinal spin-
spin correlation function (p(^ 2 )}
limit, where only the longitudinal spin-spin density matrix elements survive, the NLO
spin-spin density matrix elements are therefore not simply related to the off-diagonal
basis introduced in Ref . [5] . A discussion of the rigidity of back-to-back (tt) pairs with
respect to gluon emission in e + e~ collisions can be found in Ref. [33] .
9 Summary and conclusions
We have discussed in detail top quark polarization in above-threshold (tt) production
at a polarized linear e + e~ collider within the SM. While journeying through the three-
dimensional (P e s, cos 9, A /i)-parameter space a rich landscape of SM polarization phe-
nomena unfolds which awaits experimental confirmation or falsification. Generally speak-
ing, one needs large values of the effective beam polarization if the aim is to produce
highly polarized top quarks. Very small or zero polarization of the top quark can be
obtained by fine-tuning the parameters (P e s, v, cos 9).
The (tt)-production rate at a polarized linear e + e~ collider is governed by the gain
factor Kg = 1 — h-h + and the effective beam polarization P e g. The optimal choice as
concerns the rate is h- negative and h + positive such that one has Kg > 1 and P e ff < 0,
i.e. the optimal choice for the rate would lie in the second quadrant of the (h_, h + ) plane
in Fig. |3j The largest gain in the rate is obtained for h- = —h + = —1, i.e. for K G = 2
and P e flf = — 1-
At (Born term) threshold, one has a flat cos# distribution. As the energy increases
44
there is a quick turn into forward dominance, with little dependence on P c g. This is a
welcome feature for polarization measurements, which require large statistics and rates,
since forward production is advantageous for stable and large top quark spin effects.
More explicitly, the polarization of the top quark is generally large and more stable
against variations of the parameters P e g, cos8 and the energy in the forward region than
in the backward region.
Contrary to the rate, the polarization observables depend only on P cfT , and not sep-
arately on h_ and h + . We find that the single-spin polarization of the top quark is, in
general, strongly dependent on the effective beam polarization parameter P e g. This is
quite different for longitudinal spin-spin correlations which depend only weakly on beam
polarization effects.
In order to attain small or large values of the polarization would in general require
an extreme fine-tuning of P e g depending on cos 8 and the energy. The good news is
that the polarization properties at y/s = 500 GeV are still quite close to the polarization
properties at threshold where they are quite simple. If the aim is to achieve zero or
small polarization at = 500 GeV a choice of P e ff = 0.36 4- 0.40 leads to very small
values of \P | in the forward hemisphere where the rate is largest. At \/i = 500 GeV,
close to maximal values of the polarization \P | ~ 1 can be achieved over the whole
cos# range for effective beam polarizations close to P e ff — — 1 or P g = +1, where the
polarization is slightly larger for P eff = +1. However, a choice close to P c g = —1 is
preferred because of two reasons. First, this choice leads to larger rates and, second, the
polarization observables are more stable against variations of P e g close to P e s = — 1 than
close to P cff = +1. For ^ = 1000 GeV a total polarization of |P | > 85% and |P | > 90%
can be achieved in the forward hemisphere for P c g = — 1 and P e fr = +1, respectively.
The highest energy considered in this paper is y/s = 3000 GeV. We have found that
the polarization results at y'i = 3000 GeV are, in many aspects, not very close to their
respective asymptotic values.
For the analysis of polarization effects one also needs to know the orientation of the
polarization vector. We have given explicit results on its orientation where we have
found that, at ^ = 500 GeV, the polarization vector is still approximately aligned or
counteraligned with the electron momentum as is the case at threshold.
Our results can be viewed as a generalization of the P e g- = — 1 results of Ref. [5] to
general values of — 1 < P c g < +1. We have checked that all our Born term formulas
agree with those of Ref. [5] when we set P e g = — 1 in our Born term expressions. In
addition, we have derived simple Born term rate and polarization formulas for the case
P c ff = +1 not treated explicitly in Ref. [5]. We also provide 0(a s ) corrections to the
Born term results which we have checked against the corresponding 0(a s ) corrections in
the helicity system given in Ref. [6]. In addition, we provide radiative corrections to the
orientation angle a of the polarization vector which were not discussed in Ref. [6].
45
All the results in this paper refer to the polarization of the top quark. In order
to obtain the SM coupling predictions for the polarization of the antitop quark, let us
first set up an orthonormal spin basis for the antitop quark by replacing the momenta
in Eq. (@J by their charge conjugate partners, i.e. p t —> pi an d p e - —> Pe+- The three
orthonormal basis vectors (e^ tr \ e^ n \ e^) are now given by
g(«r) = (p e +xPf)xPf _, (ra) = p e+ x p- t ^ i d) = PL. ( 68 )
I {Pe+ X Pt) X pt | ' \p e + X pt\ ' \pt | '
In the polar angle distribution ()5]) the polar angle now refers to 9i e - and not to 9 = 6 te - ,
as in the top quark case discussed in the main part of this paper. Since the lepton pair
is back to back in the lab frame, one has Qie- — 180° — 6i e +, i-e. the two terms in Eq. fl5])
proportional to cos 6 change sign if written in terms of cos 6t e +- in the SM the rate and
the polarization components of the antitop quark are related to those of the top quark
via
at(cos9t e +) = er t (cos0 te -) ,
Pf n) (cOS^e+) = -Pf n W te -),
P} tr) (cos 6t e+ ) = P t {tr \cosd te -). (69)
As an example, and as expected, the antitop quark is predominantly produced in the
backward hemisphere relative to the e~ direction.
In polarized top decay the compositions of helicity fractions of the final state W~
bosons change relative to the helicity fractions of unpolarized top quark decay depending
on the magnitude and orientation of the polarization vector. This polarization effect has
been investigated in a number of papers where a variety of spin observables have been
defined which involve the dominant decay mode of the top quark t{\) — > b + W + (— >
i + + vi). The analysis can be done in the (e + e~) cm. frame as in Ref. [34J, in the
top quark rest frame as e.g. in Refs. [351 ESI [37], or in the W rest frame as e.g. in
Refs. [U [381 El] . References [31 [351 E3 EH] concentrate on SM predictions and discuss
radiative QCD (31 [351 EH] corrections to the respective spin observables, while Ref. [31]
analyzes the effect of non-SM interactions in the production and decay of the top quark.
The authors of Ref. [39] discuss some novel spin observables and proceed to analyze
the effect of non-SM decay vertices on these observables. QCD corrections to non-SM
interactions in the decay of an unpolarized top quark have been recently calculated in
Ref. [10]. This calculation can be easily extended to polarized top quark decay.
The discussion of this paper has focused on SM physics with longitudinal beam
polarization. Non-SM electroweak couplings on the production side, involving leptons
and quarks, and transverse beam polarization effects can be easily included using the
formalism of this paper. Transverse beam polarization effects will be discussed in a
sequel to this paper.
46
Acknowledgments
S.G. acknowledges the support by the Estonian target financed Project No. 0180056s09
and by the Deutsche Forschungsgemeinschaft (DFG) under Grant No. 436 EST 17/1/06.
B.M. acknowledges the support of the Ministry of Science and Technology of the Republic
of Croatia under Contract No. 098-0982930-2864. S.P. is supported by the Slovenian
Research Agency and by the European RTN network FLAVIAnet (Contract No. MRTN-
CT-035482).
Appendix: SM values of the
electroweak coupling coefficients
The electroweak coupling matrix elements Qij(s) needed in this paper are given by
011/12 = Q)-2Q f v e VfReXz + {v 2 e + al){v 2 f ±a))\ Xz \ 2 (=0.61/0.34),
014 = 2Q f v e a f ReXz-2(vl + a 2 e )v f a f \xz\ 2 (=-0.14),
041/42 = ZQfdeVf Re Xz - 2v e a e (vj ± a 2 f )\xz\ 2 (= -0.21/- 0.17),
#44 = -2Q f a e a f ReXz + 4v e a e v f a f \xz\ 2 ( = 0.50),
where
*'M = (.-M3 + W,r (A1)
with Mz and Tz the mass and width of the Z° and g = (16 sin 2 0w cos 2 6*vyM|) _1 =
4.229 • 10~ 5 GeV~ 2 where we have used sin 2 9w — 0.23116. Qf are the charges of the final
state quarks to which the electroweak currents directly couple; v e and a e , Vf and cif are the
electroweak vector and axial vector coupling constants. For example, in the Weinberg-
Salam model, one has v e = —1 + 4sin 2 6'vy, a e = —1 for leptons, Vf — 1 — |sin 2 ^,
cif = +1 for up-type quarks (Qf = +|), and Vf = — 1 + | sin 2 6w, a / = — 1 f° r down-type
quarks (Qf = — |). The electroweak coupling coefficients are not independent. They
satisfy the constraints
2 2\ 1 / 2
(011 ± 04l) - (014 ± 044) = 012 ± 042 • (A2)
In Eq. ( 1A1I) we have also listed the numerical values of the electroweak coefficients
for (ti) production at y'i = 500 GeV. As already mentioned in the main text, it is
safe to work in the zero width approximation for the Z boson above (tt) threshold, i.e.
we set Tz = 0. Note that the numerical values of the electroweak coefficients are only
weakly energy dependent above the (tt) threshold. The energy dependence comes from
the energy-dependent factor Xz(s) which takes the values 0.377, 0.364, and 0.352 for
yfs = 350 GeV (threshold), 500 GeV, and infinite energy, respectively.
47
For some applications it is convenient to switch to chiral representations of the initial
and final electromagnetic and weak currents as was done in Ref. [5]. Accordingly, one
defines coefficients
Ill/lr = ~Qf + (v e + a e )(v f ± a f )xz(s) . (A3)
The chiral electroweak coefficients /ll/lr can be seen to be related to the above gij via
Ill/lr = -( 5 iiT5i4-34i±544) 1/2 ( = - f .21/- 0.43) ,
hifiR = 912-942 (=0.51). (A4)
For the case P e ff = +1, one also needs the corresponding relations for the coefficients
Irr/rl = ~Qf + (v e - a e )(v f =F af)xz(s) ■ (A5)
One has
fRR/RL = -(gn±9u + 94i±94A) 1/2 (=-0.87/- 0.20),
f RR f R L = gi2 + 942 (= 0.18). (A6)
References
[1] J.E. Brau et al. [1LC Collaboration], "ILC Reference Design Report Volume 1 -
Executive Summary", arXi v:0712.1 950 [physics. acc-ph].
[2] N. Phinney, N. Toge and N. Walker, "ILC Reference Design Report Volume 3 -
Accelerator" , larXiv:0712.2361l [physics. acc-ph].
[3] M. Fischer, S. Groote, J.G. Korner, M.C. Mauser and B. Lampe,
Phys. Lett. B451 (1999) 406
[4] V. Ravindran and W.L. van Neerven, Nucl. Phys. B589 (2000) 507
[5] S. Parke and Y. Shadmi, Phys. Lett. B387 (1996) 199
[6] J. Kodaira, T. Nasuno and S.J. Parke, Phys. Rev. D59 (1998) 014023
[7] R.W. Assmann et al, CERN-2000-008, SLAC-REPRINT-2000-096
[8] V.S. Fadin and V.A. Khoze, JETP Lett. 46 (1987) 525; Yad. Fiz. 48 (1988) 487
[9] R. Harlander, M. Jezabek, J.H. Kiihn and M. Peter, Z. Phys. C73 (1997) 477
[10] M. Awramik and M. Jezabek, Acta Phys. Pol. B22 (2001) 2115
[11] G. Alexander et al, Nucl. Instrum. Meth. A610 (2009) 451
48
[12] G.A. Moortgat-Pick et al, Phys. Rept. 460 (2008) 131
[13] J.G. Kdrner and D.H. Schiller, "Helicity description of e + e~ — > qqg and e + e~ — >
QQ(1 ) — > ggg on and off the Z°: quark, gluon and beam polarization effects",
DESY Report No. 81-043, 61pp. (July 1981)
[14] S. Groote, J.G. Kdrner and M.M. Tung, Z. Phys. C74 (1997) 615
[15] F.M. Renard, "Basics Of Electron Positron Collisions", Dreux, France: Editions
Frontieres (1981) 238p.; F.M. Renard, Z. Phys. C45 (1989) 75
[16] K.I. Hikasa, Phys. Rev. D33 (1986) 3203
[17] H.E. Haber, "Spin formalism and applications to new physics searches",
arXiv:hep-ph/9405376~|
[18] J.H. Kuhn, A. Reiter and P.M. Zerwas, Nucl. Phys. B272 (1986) 560
[19] J.G. K6rner, A. Pilaftsis and M.M. Tung, Z. Phys. C63 (1994) 575
[20] S. Groote, J.G. Kdrner and M.M. Tung, Z. Phys. C70 (1996) 281
[21] S. Groote and J.G. Kdrner, Z. Phys. C72 (1996) 255;
[Erratum: Eur. Phys. J. C70 (2010) 531]
[22] S. Groote and J.G. Kdrner, Phys. Rev. D80 (2009) 034001
[23] J.H. Kuhn, T. Hahn, R. Harlander, "Top production above threshold: Electroweak
and QCD corrections combined", Talk given at SPIRES Conference C99/04/28,
arXiv:hep-ph/99122~6~2l
[24] Y.S. Tsai, Phys. Rev. D4 (1971) 2821 [Erratum-ibid. D 13 (1976) 771]
[25] J.H. Kuhn, DESY LC Notes, LC-TH-2001-004,
In "2nd ECFA/DESY Study 1998-2001", 1418-1424
[26] S. Groote, H. Liivat, I. Ots, T. Sepp, Eur. Phys. J. C66 (2010) 271
[27] S. Groote, H. Liivat, I. Ots, Nucl. Phys. B843 (2010) 213
[28] S. Groote, J.G. Kdrner and J.A. Leyva, Eur. Phys. J. C63 (2009) 391
[29] M.M. Tung, J. Bernabeu and J. Penarrocha, Phys. Lett. B418 (1998) 181
[30] S. Groote, J.G. Korner and J.A. Leyva, Phys. Lett. B418 (1998) 192
[31] A. Brandenburg, M. Flesch and P. Uwer, Phys. Rev. D59 (1998) 014001
49
[32] A. Brandenburg, M. Flesch, P. Uwer, Czech. J. Phys. 50S1 (2000) 51-58
[33] S. Grootc, J.G. K6rner and J.A. Leyva, Nucl. Phys. B527 (1998) 3
[34] B. Grzadkowski and Z. Hioki, Nucl. Phys. B585 (2000) 3
[35] A. Czarnecki, M. Jezabek, J.G. Korner and J.H. Kiihn,
Phys. Rev. Lett. 73 (1994) 384
[36] S. Groote, W.S. Huo, A. Kadeer and J. G. K6rner, Phys. Rev. D76 (2007) 014012
[37] J.G. K6rner and D. Pirjol, Phys. Rev. D60 (1999) 014021
[38] M. Fischer, S. Groote, J.G. Korner and M.C. Mauser,
Phys. Rev. D65 (2002) 054036
[39] J.A. Aguilar-Saavedra and J. Bernabeu, Nucl. Phys. B840 (2010) 349-378
[40] J. Drobnak, S. Fajfer and J.F. Kamenik, Phys. Rev. D82 (2010) 114008
50
Live Music Archive
Librivox Free Audio
Metropolitan Museum
Cleveland Museum of Art
Internet Arcade
Console Living Room
Books to Borrow
Open Library
TV News
Understanding 9/11